noone@mail.com Analysis of Dieting study 16S data % Fri Sep 7 05:46:18 2018

(1.6.1.4) Taxonomic level: otu of Subset: Patient samples

(1.6.1.4.1) Loading counts and metadata

Subreport

(1.6.1.4.2) Data analysis

(1.6.1.4.2.1) Richness and diversity estimates Before count filtering

Subreport

Filtering abundance matrix with arguments [ min_mean :10, min_quant_incidence_frac:0.25, min_quant_mean_frac :0.25]. Filtering features

Note that some community richness estimators will not work correctly if provided with abundance-filtered counts

After filtering, left 67 records for 113 features

Wrote counts and metadata for raw counts After final feature filtering to files data/1.6.1.4.2.1-3233b4c0ccdsamples.raw.16s.l.otu.count.tsv,data/1.6.1.4.2.1-3233b4c0ccdsamples.raw.16s.l.otu.attr.tsv

Wrote counts and metadata for proportions counts After final feature filtering to files data/1.6.1.4.2.1-32320999416samples.proportions.16s.l.otu.count.tsv,data/1.6.1.4.2.1-32320999416samples.proportions.16s.l.otu.attr.tsv

(1.6.1.4.2.2) DESeq2 tests and data normalization

Love MI, Huber W, Anders S (2014). “Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2.” Genome Biology, 15, 550. doi: 10.1186/s13059-014-0550-8 (URL: http://doi.org/10.1186/s13059-014-0550-8).

(1.6.1.4.2.2.1) Table 445. DESeq2 results for task: visit;[ alpha:0.05]. Full dataset is also saved in a delimited text file (click to download and open e.g. in Excel) data/1.6.1.4.2.2.1-323604c7d0a.1.6.1.4.2.2.1.a.nam.csv

(1.6.1.4.2.2.1) Figure 1028. Clustered heatmap of normalized abundance values. Number of cluster splits is determined automatically with method fpc::pamk. Image file: plots/323ed79c0a.svg.

(1.6.1.4.2.2.1) G-test of independence between automatic cluster splits and attribute 'visit'. Number of cluster splits is determined automatically with method fpc::pamk.

Wrote counts and metadata for raw counts Data used for heatmap with added row cluster splits to files data/1.6.1.4.2.2.1-323717f952csamples.raw.htmap.count.tsv,data/1.6.1.4.2.2.1-323717f952csamples.raw.htmap.attr.tsv

(1.6.1.4.2.3) Default transformations for further data analysis

Specific methods can override these and use their own normalization.

Count normalization method for data analysis (unless modified by specific methods) : [ drop.features:List of 1, ..$ :"other", method.args :List of 1, ..$ theta:1, method :"norm.ihs.prop"]

Wrote counts and metadata for raw counts Normalized after default transformations to files data/1.6.1.4.2.3-32319dd7d29samples.raw.16s.l.otu.count.tsv,data/1.6.1.4.2.3-32319dd7d29samples.raw.16s.l.otu.attr.tsv

(1.6.1.4.3) Stability selection analysis for response ( visit )

(1.6.1.4.3) Summary of response variable visit.

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  1.000   1.000   2.000   2.254   3.000   4.000 

Hofner B, Hothorn T (2017). stabs: Stability Selection with Error Control. R package version 0.6-3, <URL: https://CRAN.R-project.org/package=stabs>.

Hofner B, Boccuto L, Göker M (2015). “Controlling false discoveries in high-dimensional situations: Boosting with stability selection.” BMC Bioinformatics, 16, 144. <URL: http://dx.doi.org/10.1186/s12859-015-0575-3>.

Thomas J, Mayr A, Bischl B, Schmid M, Smith A, Hofner B (2017). “Gradient boosting for distributional regression - faster tuning and improved variable selection via noncyclical updates.” Statistics and Computing. doi: 10.1007/s11222-017-9754-6 (URL: http://doi.org/10.1007/s11222-017-9754-6), Online First.

This multivariate feature selection method implements stability selection procedure by Meinshausen and Buehlmann (2010) and the improved error bounds by Shah and Samworth (2013). The features (e.g. taxonomic features) are ranked according to their probability to be selected by models built on multiple random subsamples of the input dataset.

Base selection method parameters that were chosen based on cross-validation are: [ alpha:0.3]

All base selection method parameters are: [ family :"gaussian", standardize:TRUE, alpha :0.3]

Stability selection method parameters are: [ PFER :0.05, sampling.type:"SS", assumption :"r-concave", B :400, q :8]

(1.6.1.4.3) Table 446. Selection probability for the variables. Probability cutoff=0.845 corresponds to per family error rate PFER=0.0498. Full dataset is also saved in a delimited text file (click to download and open e.g. in Excel) data/1.6.1.4.3-3234ec4c761.1.6.1.4.3.a.name.ta.csv

(1.6.1.4.3) Figure 1029. Selection probability for the top ranked variables. Probability cutoff=0.845 corresponds to per family error rate PFER=0.0498 (vertical line). Image file: plots/323578100fc.svg.

(1.6.1.4.4) PermANOVA (adonis) analysis of normalized counts

Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, Minchin PR, O'Hara RB, Simpson GL, Solymos P, Stevens MHH, Szoecs E, Wagner H (2018). vegan: Community Ecology Package. R package version 2.5-2, <URL: https://CRAN.R-project.org/package=vegan>.

Non-parametric multivariate test for association between normalized counts and meta-data variables. Dissimilarity index is euclidean.

(1.6.1.4.4) Association with visit paired by subject with formula count~visit with strata = SubjectID.

Permutation test for adonis under reduced model
Terms added sequentially (first to last)
Blocks:  m_a$attr[, strata] 
Permutation: free
Number of permutations: 4000

adonis2(formula = as.formula(formula_str), data = m_a$attr, permutations = perm, method = dist.metr)
         Df SumOfSqs      R2      F Pr(>F)
visit     1    0.088 0.01223 0.8045 0.7248
Residual 65    7.110 0.98777              
Total    66    7.198 1.00000              

(1.6.1.4.4) Association with visit paired by subject Adonis summary.

Count normalization method for abundance plots : [ drop.features:List of 1, ..$ :"other", method.args : list(), method :"norm.prop"]

(1.6.1.4.5) Plots of Abundance.

Subreport

(1.6.1.4.5) Figure 1056. Clustered heatmap of normalized abundance values. Number of cluster splits is determined automatically with method fpc::pamk. Image file: plots/3231e6a0de1.svg.

(1.6.1.4.5) G-test of independence between automatic cluster splits and attribute 'visit'. Number of cluster splits is determined automatically with method fpc::pamk.

person) (????). morpheus: Interactive heat maps using 'morpheus.js' and 'htmlwidgets'. R package version 0.1.1.1, <URL: https://github.com/cmap/morpheus.R>.

(1.6.1.4.5) Widget 462. Dynamic Morpheus heatmap of normalized abundance values. It is available here through the link only because it can take a while to render for large datasets. This is very customizable. What you will see initially is just a default starting configuration. Explore its menus. Click to see HTML widget file in full window: ./1.6.1.4.5-32350c78094Dynamic.Morpheus.hea.html

Wrote counts and metadata for raw counts Data used for heatmap with added row cluster splits (clustering by abundance profile) to files data/1.6.1.4.5-32331812ce8samples.raw.htmap.count.tsv,data/1.6.1.4.5-32331812ce8samples.raw.htmap.attr.tsv

(1.6.1.4.5) Figure 1057. Clustered heatmap of diversity and normalized abundance values. Number of cluster splits is determined automatically with method fpc::pamk. Image file: plots/32379f9d705.svg.

(1.6.1.4.5) G-test of independence between automatic cluster splits and attribute 'visit'. Number of cluster splits is determined automatically with method fpc::pamk.

Wrote counts and metadata for raw counts Data used for heatmap with added row cluster splits (clustering by Renyi diversity indices) to files data/1.6.1.4.5-3235f3c374asamples.raw.htmap.count.tsv,data/1.6.1.4.5-3235f3c374asamples.raw.htmap.attr.tsv

(1.6.1.4.6) Ordinations,

Subreport

(1.6.1.4.7) Network Analysis Feature correlation with overlaid differential abundance results

Subreport

Design: Rapporter Team © 2011-2014 | Backend: pander | License: AGPL3 or OSL | Styled with skeleton